@inproceedings{li-etal-2025-formalizing,
title = "Formalizing Test-Time Compute for Function-Level Code Generation",
author = "Li, Haau-Sing and
Fernandes, Patrick and
Gurevych, Iryna and
Martins, Andre",
editor = "Inui, Kentaro and
Sakti, Sakriani and
Wang, Haofen and
Wong, Derek F. and
Bhattacharyya, Pushpak and
Banerjee, Biplab and
Ekbal, Asif and
Chakraborty, Tanmoy and
Singh, Dhirendra Pratap",
booktitle = "Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics",
month = dec,
year = "2025",
address = "Mumbai, India",
publisher = "The Asian Federation of Natural Language Processing and The Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-ijcnlp.70/",
pages = "1157--1170",
ISBN = "979-8-89176-303-6",
abstract = "Test-time compute has emerged as a powerful paradigm in function-level code generation. However, previous proposed strategies have been viewed as disparate, thus lacking a fair apples-to-apples analysis enabling understanding of their operational mechanisms in execution-based benchmarks. Therefore, we present a mathematical framework that unifies generation and reranking with theoretical justifications through the lens of Minimum Bayes Risk (MBR) decoding. Our proposed framework leads to key research questions regarding the effectiveness of using parallel and/or iterative sampling, design choices of reranking signals and soft/hard MBR utility functions, and behaviors of the final selected program across different methods. Our empirical findings highlight the importance of the diversity of sampled candidates (over self-improvement), reranking with simple and high-quality signals, and the effectiveness of test-time compute to select programs that manifest general and edge test case robustness. We will open-source our analysis toolkit and implementation to enable reproducible research.We open-source our analysis toolkit and implementation to enable reproducible research."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2025-formalizing">
<titleInfo>
<title>Formalizing Test-Time Compute for Function-Level Code Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Haau-Sing</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Patrick</namePart>
<namePart type="family">Fernandes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iryna</namePart>
<namePart type="family">Gurevych</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haofen</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Derek</namePart>
<namePart type="given">F</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biplab</namePart>
<namePart type="family">Banerjee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asif</namePart>
<namePart type="family">Ekbal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhirendra</namePart>
<namePart type="given">Pratap</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>The Asian Federation of Natural Language Processing and The Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mumbai, India</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-303-6</identifier>
</relatedItem>
<abstract>Test-time compute has emerged as a powerful paradigm in function-level code generation. However, previous proposed strategies have been viewed as disparate, thus lacking a fair apples-to-apples analysis enabling understanding of their operational mechanisms in execution-based benchmarks. Therefore, we present a mathematical framework that unifies generation and reranking with theoretical justifications through the lens of Minimum Bayes Risk (MBR) decoding. Our proposed framework leads to key research questions regarding the effectiveness of using parallel and/or iterative sampling, design choices of reranking signals and soft/hard MBR utility functions, and behaviors of the final selected program across different methods. Our empirical findings highlight the importance of the diversity of sampled candidates (over self-improvement), reranking with simple and high-quality signals, and the effectiveness of test-time compute to select programs that manifest general and edge test case robustness. We will open-source our analysis toolkit and implementation to enable reproducible research.We open-source our analysis toolkit and implementation to enable reproducible research.</abstract>
<identifier type="citekey">li-etal-2025-formalizing</identifier>
<location>
<url>https://aclanthology.org/2025.findings-ijcnlp.70/</url>
</location>
<part>
<date>2025-12</date>
<extent unit="page">
<start>1157</start>
<end>1170</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Formalizing Test-Time Compute for Function-Level Code Generation
%A Li, Haau-Sing
%A Fernandes, Patrick
%A Gurevych, Iryna
%A Martins, Andre
%Y Inui, Kentaro
%Y Sakti, Sakriani
%Y Wang, Haofen
%Y Wong, Derek F.
%Y Bhattacharyya, Pushpak
%Y Banerjee, Biplab
%Y Ekbal, Asif
%Y Chakraborty, Tanmoy
%Y Singh, Dhirendra Pratap
%S Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics
%D 2025
%8 December
%I The Asian Federation of Natural Language Processing and The Association for Computational Linguistics
%C Mumbai, India
%@ 979-8-89176-303-6
%F li-etal-2025-formalizing
%X Test-time compute has emerged as a powerful paradigm in function-level code generation. However, previous proposed strategies have been viewed as disparate, thus lacking a fair apples-to-apples analysis enabling understanding of their operational mechanisms in execution-based benchmarks. Therefore, we present a mathematical framework that unifies generation and reranking with theoretical justifications through the lens of Minimum Bayes Risk (MBR) decoding. Our proposed framework leads to key research questions regarding the effectiveness of using parallel and/or iterative sampling, design choices of reranking signals and soft/hard MBR utility functions, and behaviors of the final selected program across different methods. Our empirical findings highlight the importance of the diversity of sampled candidates (over self-improvement), reranking with simple and high-quality signals, and the effectiveness of test-time compute to select programs that manifest general and edge test case robustness. We will open-source our analysis toolkit and implementation to enable reproducible research.We open-source our analysis toolkit and implementation to enable reproducible research.
%U https://aclanthology.org/2025.findings-ijcnlp.70/
%P 1157-1170
Markdown (Informal)
[Formalizing Test-Time Compute for Function-Level Code Generation](https://aclanthology.org/2025.findings-ijcnlp.70/) (Li et al., Findings 2025)
ACL
- Haau-Sing Li, Patrick Fernandes, Iryna Gurevych, and Andre Martins. 2025. Formalizing Test-Time Compute for Function-Level Code Generation. In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 1157–1170, Mumbai, India. The Asian Federation of Natural Language Processing and The Association for Computational Linguistics.